Documente Academic
Documente Profesional
Documente Cultură
By
MARK APPLETON
University of Queensland
October 2001
i
73 Hawken Dr.
Head of School;
University of Queensland
In accordance with the requirements of the degree of Bachelor of Engineering in the division of
Electrical and Engineering, I present the following thesis entitled “An Investigation of the Interface
Between Various Overhead Distribution Insulator Types and 11kV Covered Conductor”. This
work was sponsored by Energex and completed under the supervision of Dr Tapan Saha.
I declare that the work submitted in this thesis is my own work, except as acknowledged in the
text, and has not been previously submitted for a degree at the University of Queensland or any
other institution.
Yours faithfully,
Mark Appleton
ii
Acknowledgements
year. He has been most helpful in lending his expertise and knowledge and ensuring that my
work progressed with as few problems as possible. When problems arose, he was willing to
The High Voltage Laboratory supervisor, Mr. Steven Wright, is also worthy of many thanks. He
was responsible for ensuring safety procedures were abided by and that proper testing
c
Mr. John M Donald made many sacrifices through out the year to help with the partial discharge
testing and he was always available for consultation about any problems I encountered.
Energex has been extremely helpful with regards to my thesis. My Energex supervisors, Mr. Pat
Pearl and Mr. Greg Dowling, were both willing to take time out of their busy schedule to assist me
when required. They were responsible for providing and delivering the equipment punctually. I
would also like to thank Mr. Shane Bayley for providing the CAD designs of the equipment.
Finally I would like to thank all my friends for being supportive; especially Julian, Lachlan and
David for helping with the testing even though they were all extremely busy as well.
iii
Abstract
This thesis details the testing of different insulator configurations for use with 11kV Covered
Conductor Thick distribution lines. An energy distribution company in South East Queensland
called Energex are interested in implementing covered conductors as a means to reducing the
number of vegetation and wildlife related outages. The insulators investigated were the Pin Post,
Clamp Top and the Tie Top. This work also attempted to determine the impact and usefulness of
stripping the conductor near the attachment of the insulators. As partial discharges are
proportional to the surrounding electric field, the field was modeled by finite element analysis and
measured in the laboratory. The leakage current of the each configuration was measured so as
It was found that the Covered Conductor Thick passed the Australian Standards for partial
discharges and even performed better than the stripped conductor in most cases. The partial
discharges detected within the conductor were of the same magnitude as the background noise
and were thus so small that it seems superfluous to strip at 11kV, though at increased operating
voltages the discharges may be larger and it may then become necessary. When the conductor
and insulators were modeled the stripped conductor always had a smaller electric field present
than the covered conductor did. This was supported by the laboratory measurements that also
showed that the pin post insulator had the greatest electric field for all samples of condutor.
iv
Contents
ABSTRACT ......................................................................................................................... IV
CHAPTER 3: T HEORY...........................................................................................................8
REFERENCES ..................................................................................................................... 63
BIBLIOGRAPHY .................................................................................................................. 65
v
APPENDIX A – EMF RES ULTS ............................................................................................ 66
vi
List of Figures
FIGURE 4-1: THE DIFFERENT INTERNAL LAYERS OF COVERED CONDUCTOR THICK. [8]
............................................................................................................................................ 11
FIGURE 4-3: THE TRIDENT STRUCTURE WITH TWO TIE TOP INSULATORS AND A CLAMP
TOP INSULATOR. ............................................................................................................... 13
FIGURE 4-4: THE SHORT CROSS ARM WITH TWO PIN POST INSULATORS ATTACHED. . 14
FIGURE 4-5: THE THREE TYPES OF INSULATORS. THE CLAMP TOP (LEFT), THE TIE TOP
(CENTRE) AND THE PIN POST (RIGHT). ............................................................................. 15
FIGURE 4-6: THE ERA DISCHARGE DISPLAY. THE DISCHARGES ARE DI SPLAYED ON
THE GREEN OSCILLOSCOPE. ............................................................................................ 16
FIGURE 4-7: THE GAUSS-MAUS. THE DETECTOR IS ON THE LEFT AND THE DISPLAY IS
ON THE RIGHT.................................................................................................................... 17
FIGURE 5-1: LEAKAGE CURRENT CIRCUIT FOR A) PIN POST INSULATOR AND B) CLAMP
AND TIE TOP INSULATOR. ................................................................................................. 19
FIGURE 5-4: OSCILLOSCOPE OUTPUT WITH NEGATIVE PEAK CORONA AT THE TOP OF
THE ELLIPSE AND THE 100V INPUT AT THE BOTTOM RIGHT CORNER. ........................... 22
FIGURE 5-5: THE EIGHT POSITIONS USED FOR MEASURING THE ELECTRIC FIELD. ...... 24
FIGURE 5-6: A TIE TOP INSULATOR DRAWN USING A SPACING OF 1 (LEFT) AND 50
(RIGHT). ................................ ................................ ................................ .............................. 26
FIGURE 6-2: AVERAGE EMF MEASUREMENTS FOR THE CLAMP TOP INSULATOR ......... 34
FIGURE 6-3: AVERAGE EMF MEASUREMENTS FOR THE SIDE TIE INSULATOR............... 35
FIGURE 6-4: AVERAGE EMF MEASUREMENTS FOR THE PIN POST INSULATOR ............. 35
vii
FIGURE 6-5: PARTIAL DISCHARGE CLAMP TOP - COVERED CONDUCTOR THICK .......... 39
FIGURE 6-9: PARTIAL DISCHARGE PIN POST - COVERED CONDUCTOR THICK .............. 41
FIGURE 6-14: PARTIAL DISCHARGE SIDE TIE ASCENDING - COVERED VS STRIPPED .... 44
viii
Chapter 1
Introduction
Throughout the world, many distribution companies are turning to covered conductors as a
means to reducing the number of faults occurring along their transmission lines. These covered
conductors reduce the chance of faults caused by nearby vegetation or wildlife making contact
with one or more of the phases, or two or more of the phases clashing together. However,
covered conductors do have limitations. The currently used pin post insulators, which are
attached between the conductor and the power poles, have a large metal pin that holds the
insulator in place. This pin is at a much lower voltage than the nearby conductor is and therefore
a large electric field is present. The pin post insulators also have a higher dielectric constant than
the covering. This difference in dielectrics causes a majority of the electric field to be distributed
across the covering. This concentration of electric field directly affects the amount and the
frequency of partial discharges. Over time, these partial discharges can cause such extensive
damage to the covering and the conductor that the conductor snaps. This is a significant problem,
as this poses both a major health risk and can disrupt power from being delivered to the
consumer.
implementing these covered conductors. One of the main objectives of Energex is to provide a
regular service to their customers with as few interruptions as possible. Previously in Brazil there
has been some success in remedying the problem of coronal discharge in covered conductors by
stripping the covering from the conductor near where the line is attached to the pin insulator.
Energex therefore performed tests on the effectiveness of stripping the covered conductors. They
discovered that not only was this stripping difficult and time consuming, but also the stripping
tools often broke. This forced them to try and find a different solution.
1
This thesis is concerned with the interactions between the covered conductor and a group of
different insulators that Energex is considering using in the future or is using at present. By
modelling the electric fields and then by measuring these fields as well as the size of partial
discharges produced in the conductors and the amount of leakage current, the insulators and
configurations were able to be compared. Finite Element Method analysis utilising the QuickField
program was used to model the electric field expected in the surrounding area. As the version
used was an evaluation version, the accuracy was reduced, though a good qualitative analysis
was still achieved. The measurements of the electric field were then performed at various
locations with a Gauss-Maus. The partial discharges were measured with the ERA Discharge
By performing these tests it is hoped to determine which of the available insulators is best suited
to being used with the covered conductor and whether or not the covered conductors require
stripping. This thesis also demonstrates valid testing techniques for covered conductors and with
some slight adaptations these techniques could be used for different voltage lines.
A brief outline of the contents of this thesis follows. An overview of useful background material is
initially investigated. This section is then followed by useful theoretical definitions and
description of the experimental procedure undertaken is provided, followed by the results to these
experiments and the conclusions that can be drawn from these results.
2
Chapter 2
Literature Review
This chapter provides background information that is relevant to this thesis topic. Firstly, the
operation of covered conductors is described. This section also provides some theories for
reducing the chance of conductor damage. Followed then are reasons for investing in covered
conductors and the rationale for Energex’s decision to use Covered Conductor Thick.
Covered conductors serve to provide a more reliable power supply to the public. They do this by
preventing short circuits between the conductors and the external environment. In Scandinavia
many tests have been performed on the cables. Trees were felled onto lines and they were left
there for up to a year. Audible partial corona discharges occurred, but this discharge was found to
have very little effect on the mechanical or electrical properties of the line, even after such a long
time. Usually when one of the phases of a feeder is grounded, the circuit breakers are opened by
the relay, causing the power flow to the feeder is stopped. The power to the feeder in a covered
conductor system does not need to be removed in such a case, as the electrical properties are
unaffected. This provides a more reliable supply to the consumer. The implementation of these
cables seems like an ideal solution, but there can be many problems involved with them. [1]
One major problem with the covered conductor occurs when lightning strikes. The flashover
produced occurs between the support points, much as in bare conductors, but in covered
conductors the arcing is not distributed along a large length, but in a small puncture point. Any
subsequent flashovers will occur at this point and the aluminium strands will rapidly deteriorate.
This deterioration reduces the mechanical properties of the line and can cause the cable to snap,
3
There are a number of different countermeasures that can reduce the risk of burndown. These
countermeasures include the use of high speed interrupting relays; increasing the Basic
Insulation Level (BIL) of the conductor; stripping the covering from the conductor in the region
near the post attachment; using surge arresters; and using insulator and insulator tie
configurations that reduce the electric field in the nearby area. In the near future, it is unlikely that
high-speed relays will be able to clear faults quickly enough to prevent burndown. This is because
it is doubtful whether there are available switches that will be able to operate at the required
speed, which is approximately 1–2 cycles. As lightning induced overvoltages easily exceed the
impulse strength of the circuit, with or without coverings on the conductors, increasing the Basic
Insulation Level of the covering, to reduce the likelihood of flashover, would not be a viable
economic possibility. Surface leakage current and partial discharge have been linked to possible
causes of conductor burndown. Partial discharges are discharges that do not fully cross the
electrodes, in this case the two sides of a cavity within the polyethylene covering of the
conductor. Over time these discharges can increase in size to a dangerous magnitude and cause
damage to the covered conductor. Therefore to reduce the electric field producing these, the
covered conductor may be stripped in the insulator tie region. [2] [4]
It is unsure whether stripping actually reduces the effect of partial discharge and may not have an
impact on reducing the chances of burndown. In fact, when there is a flashover to the stripped
region, the arc comes to rest at the start of the covering, which can have an extreme heating
The disadvantages to stripping are that the actual task is especially difficult and time consuming.
Work teams have to be trained in the use of the stripping tools and new tools will have to be
purchased. Another problem is that there is now an increased possibility of wildlife related
outages, due to the live conductor exposed near the post insulators. [5]
4
Stripping can possibly be avoided with the strategic placement of surge arresters. The frequency
of these surge arresters is dependent on the likelihood of lightning strikes in the surrounding area.
In dense trees, the probability of a strike is much less than in open country, therefore the surge
arresters could be used less frequently. The use of surge arresters does not totally prevent the
risk of flashover and burndown, but they can reduce it to an acceptable level. Surge arresters are
fairly expensive, typically a few hundred dollars, and over an entire network this is a considerable
During 1983 bushfires swept through South Australia and caused extensive damage and tragic
loss of life. The cause of this tragic incident is believed to be bare conductors. This forced the
Electricity Trust of South Australia (ETSA) to investigate options of reducing the dangers of
bushfire. In 1987 ETSA implemented an Aerial Bundled Cable (ABC) system which was later
abandoned in favour of either Insulated Unscreened Conductor (IUC – which is identical to CCT)
or Covered Conductor (CC). These covered systems greatly reduce the chance of transmission
Apart from the electrical and mechanical benefits of the covered conductor, there are aesthetic
benefits as well. In a non-covered system when two phases clash a phase-to-phase fault occurs,
so the phases are spaced apart. This is not a problem in a covered conductor situation, so the
phases can be situated closer together. Landowners have reacted very positively and so have
ground crews who now have to clear smaller paths through dense bushland, as wildlife and fallen
Energex investigated many possibilities before settling on using Covered Conductor Thick. A
study was performed analysing the benefits and drawbacks of using the available options. The
options investigated were Covered Conductor (CC), Covered Conductor Thick (CCT), Non-
metallic Screened Aerial Bundled Cable (NMSHVABC), Metallic Screened Aerial Bundled Cable
5
CC and CCT are very similar. Both are single-core aerial conductors covered with Cross Linked
Polyethylene (XLPE), but CCT also has an outer covering of High Density Polyethylene (HDPE).
CC is able to withstand occasional contact with other phases or trees, while the extra outer
NMSHVABC is a bundled conductor comprising of three insulated cores and a bare support and
earthing conductor. The innermost covering of the three phases is made of a conductor screen,
then a XLPE covering, another insulating screen and then an optional HDPE sheath. The support
conductor also provides earthing for the outermost semiconductive layer. The MSHVABC is
similar to the NMSHVABC but has an optional water-swellable tape, a copper wire screen and a
separator tape between the second insulation screen and the HDPE exterior.
The Spacer system involves three insulated phases supported from an aluminium alloy catenary
wire by HDPE spacers. The covering of the conductors is constructed of an inner layer of Low
Density Polyethylene (LDPE) and an outer layer of black or grey HDPE. Unlike the other systems,
Spacer is not water blocked, though no problems of water corrosion have been reported.
Aerial Spacer Cable had the least similarity with the present systems and was not investigated as
fully as the systems that possess the greatest similarities. This is because the other systems
would be able to be retrofitted to existing equipment and would require less training for staff.
MSHVABC was also rejected, as it is approximately twice the weight and price of the non-metallic
type and also has lower normal and fault current ratings. The remaining possibilities were
compared with the standard Bare Conductor as well as with Underground Cable.
It was found that there was not a clear economic winner, therefore Energex proposed two
schemes. The first would be to use CCT on all lines. The second option was to use NMSHVABC
for new lines in timbered areas and CC for new lines in open areas. The second option was found
6
to be slightly better economically, but posed other problems in that NMSHVABC requires more
expensive equipment and training than CCT and is also not suitable for live line work. CC is
cheaper than CCT but its main advantage is that it is better at preventing the ignition of bushfires.
This is more important in the southern states of Australia as they have dry summers, while in
Queensland, the chance of bushfire is reduced due to wetter summers. Reduced tree trimming is
not allowed for CC and after a storm the lines need to be patrolled to check for branches lying
across the lines. This is not the case for CCT, which is able to withstand entire trees lying across
the lines for months at a time. The final advantage to CCT is that as only one type of cable is
The final decision to be made was what cross sectional area cable to use. Again there were two
2 2 2
options, use 120mm cable everywhere or use 80mm with 180mm . The second option was
found to have a cost advantage of only 6.5%, but would provide problems again with storage
2
space. Therefore only the 120mm cable was chosen. [8]
This background information provides an insight into why many distribution companies are
planning on implementing covered conductors. Not only are covered conductors capable of
providing power with a decrease in the amount of outages, they also require less tree trimming
and are therefore more aesthetically pleasing and less harmful to the environment. Certain
problems were mentioned regarding damage to the conductor caused by lightning and partial
discharges. One of the possible solutions presented was the stripping of the insulating covering
from the conductor near the insulator tie region. This thesis investigates whether the stripping of
the conductor actually reduces the electric field and whether the amount and size of partial
7
Chapter 3
Theory
A partial discharge is an electric discharge that does not completely bridge the electrodes. These
discharges are generally small, but because of their repetitive nature they can cause progressive
deterioration that may lead to failure in the covered conductor. Therefore it is essential to be able
Coronal discharges are a type of partial discharges that usually occur at sharp edges or points,
as this is where the electric field is the greatest. These discharges can be detrimental to partial
discharge detection. When the potential is increased, a set of lines will appear on the oscilloscope
display that do not increase in magnitude, but will increase in quantity. The harmful side of
coronal discharges is that all sharp edges and points must be removed from the test circuit, as
these will produce discharges that will be detected by the partial discharge detector. This makes
the construction of the circuit time consum ing, as it is imperative to use metal rods as conductors
and hollow metal spheres to ensure that a round object surrounds all edges. [4]
This thesis is concerned with the partial breakdown that occurs within the covering of 11kV
distribution lines. This can occur in a manner of ways including, internal cavities parallel and
perpendicular to the electric field, in a spherical cavity and via treeing. These partial discharges if
large enough can significantly damage the covering and the conductor and can, given enough
time, cause the conductor to snap and fall down. When the conductor snaps, it produces a health
risk if the fault is not detected and power disabled. If this occurs, the conductor is still live and
could be fatal if touched. Though this is harder to detect than when a bare conductor snaps, it is
still safer as the covering still provides protection from the live conductor.
8
Figure 3-1: i) Internal discharges, ii) Internal Discharges – a) perpendicular to the electric
field, b) spherical cavity, c) parallel to the electric field and d) situated in a longitudinal
field, iii) Surface discharges, iv) Corona discharges and v) Discharges in electrical trees.
[4]
9
The electric field surrounding the insulator is extremely important, as it determines the magnitude
of the effect of any partial discharge that may occur in the covered conductor. The electric field is
the derivative of the voltage with respect to distance. Therefore when there is a large potential
10
Chapter 4
Apparatus
This chapter describes all of the equipment used throughout the testing of the different insulator
and conductor configurations. This is done to outline the differences in each of the test specimens
4.1 Conductors
The covered conductor tested was an All Aluminium Alloy (AAAC) core made of aluminium alloy
1120. To reduce the amount of storage space, ENERGEX decided to implement only one size
2
and type of conductor, the 120mm made of 7 x 2.75mm diameter strands. It was found that this
type of conductor was the most versatile and cost effective. It has an inner insulation of cross-
Figure 4-1: The Different Internal Layers of Covered Conductor Thick. [8]
11
4.1.2 Stripped Conductor
The stripped conductor is identical to the covered conductor except that it has the insulation
covering stripped near the insulator attachments. This is hoped to reduce the electric field present
in the region. Reducing the electric field is hoped to decrease the amount and size of partial
This bare conductor is currently in extensive use with the pin type insulators placed.
Figure 4-2: The Three Types of Conductor Tested. Covered Conductor Thick (top),
12
4.2 Supporting Structures
The Trident structure is becoming one of Energex’s most common structures. It has one vertical
insulator and two insulators attached to a crosspiece at 78? to the vertical. The mounting is made
of metal and is usually attached to a wooden or concrete pole approximately ten metres above
the ground. In the tests performed, the conductor was held 2.4m from the ground as a larger pole
was unavailable and would not have caused any significant difference in the results. It would also
have proved difficult to attach each of the components and take measurements.
To enable this structure to travel around a bend, it is standard Energex procedure to replace one
of the Tie Top insulators with a Clamp Top insulator. Even when the trident structure is used as a
corner, the difference in structure has very little effect on the Basic Insulation Level.
Figure 4-3: The Trident Structure with two Tie Top insulators and a Clamp Top insulator.
13
4.2.2 Short Cross Arm
The cross arm used during the testing resembles the cross arms used by Energex for attaching
pin post insulators when the conductors are travelling in a straight line. The cross arm is not as
long as the type used by Energex, but is identical in all other respects. This difference in length is
acceptable due to the fact that only one phase is being energised at a time and therefore the
distance between the insulators is not important. The pin post insulators were placed 0.15m from
Figure 4-4: The Short Cross Arm with two Pin Post insulators attached.
4.3 Insulators
The tie top insulator is the most common insulator used on the Trident structure. The conductor is
tied onto the insulator with a polyethylene covered insulator tie. This type of tie is moulded in a
set position and comes in two shapes, the top tie and the side tie. The top tie is used on the
middle insulator when the conductors are going straight ahead and on the outside insulator when
the structure is turning around a corner. The side tie is used on the side insulators when the
conductor is going straight ahead and on the middle insulator when the conductor is travelling
around a bend. In this thesis this insulator is sometimes referred to as the Side Tie.
14
4.3.2 Clamp Top Insulator
The Clamp Top insulator is used on the Trident structure to bend around corners, where it is
placed on the inside of the curve. The Clamp Top insulator does not require any insulator ties to
attach the conductor to the insulator, as the conductor is held in place by the clamp, which is
The Pin Post insulator is a much smaller insulator then the Clamp Top or Tie Top insulators. It
has small porcelain head attached to an eight-inch metal pin that is used to attach the insulator
onto the cross arm. When this type of insulator is used with CCT, it is suspected of increasing
the electric field around the insulator and therefore causing larger and more frequent partial
discharges. To verify this theory, experiments were performed to compare the attributes of the
discharges detected with bare, completely covered and partially stripped conductors.
Figure 4-5: The three types of insulators. The Clamp Top (left), the Tie Top (centre) and the
15
4.4 ERA Partial Discharge Display
This partial discharge detector has a range of different detecting units. These units are used for
measuring varying sized discharges. The smaller units provide greater accuracy for small
discharges and the larger units should be used for large discharges. As the discharges being
detected are very small in magnitude the smallest detecting unit is being utilised. This unit
Once the discharge is detected, it is displayed by an oscilloscope on the ERA Discharge Display.
Figure 4-6: The ERA Discharge Display. The discharges are displayed on the green
oscilloscope.
16
4.5 Gauss-Maus
The Gauss-Maus was used to measure the electric field present surrounding the conductor when
it was energised at rated voltage. There are various other tools capable of doing this on the
market, though the Gauss-Maus has a long connection between the detector and the display.
This enabled the detector to be placed close to the conductor, while the display was read behind
a safety screen.
Figure 4-7: The Gauss-Maus. The detector is on the Left and the disp lay is on the right.
17
Chapter 5
Experimental Procedure
Each of the following three tests were performed upon the different insulator and conductor
configuration, EMF measurements were taken with four samples of conductor per insulator while
partial discharge measurements were taken for two samples per insulator. Covered Conductor
Thick and stripped conductor were tested with all three types of insulator, while the bare
One of the comparisons between the different arrangements is the amount of leakage current
flowing across the outside of the insulators. The insulators are designed to maximise this distance
by having folds along the outside, which increases the impedance thus reducing the magnitude of
the current. The leakage current should be reduced to as little as possible, as it is a loss to the
system. If this loss can be reduced, the power system becomes more efficient and therefore less
costly to the distributor. The leakage current should be smaller for the covered conductor than the
bare conductor as the polyethylene covering offers extra insulation. The bare conductor and the
stripped conductor should have approximately the same leakage current, as both of these
There are no Australian Standards recommending test procedure or acceptable values, so a new
test was designed and a straight comparison between conductor and insulator combinations was
performed. The leakage current was measured by raising the conductor to 5.25kV, 6.35kV and
7.57kV. These values were chosen arbitrarily to give an indication of how leakage current varies
18
with voltage. Th en the current was measured through an ammeter that was connected from the
base of the pin for the Pin Post insulator and from the base of the Trident structure for the Clamp
Figure 5-1: Leakage Current Circuit for a) Pin Post Insulator and b) Clamp and Tie Top
Insulator.
19
5.2 Partial Discharge
The next test was to determine the onset of partial discharges in one phase of the system. This is
important, because if partial discharges begin occurring at a voltage above the rated voltage, they
are less likely to occur when implemented and are therefore less likely to damage the covering
and the conductor. There is still the possibility of partial discharges occurring though, as the
inception voltage is higher than the retaining voltage. Therefore if the inception voltage is only
slightly higher than the rated voltage, and the retaining voltage is below the rated voltage, a small
voltage swell could start the partial discharges and they would continue to occur as long as the
retaining voltage is exceeded. Only one phase of the system could be tested at a time, as the
The relevant Australian Standard for the detection of partial discharges in a non-metallic
screened conductor is AS3599.2. [9] This standard provides guidelines for testing procedure and
recommends that the maximum discharges on a conductor at 10kV (150% rated voltage) and at
13kV (200% rated voltage) should be 5pC and 50pC respectively. If these values are exceeded,
the conductor is more likely to be susceptible to damage, and over a prolonged period of time, to
snapping. It was also decided to measure the discharges at rated voltage (6.35kV).
Partial discharges can be detected with the ERA Partial Discharge Display in a number of
different methods that require different circuits. The most common methods are straight detection
Balanced detection uses impedances that can be varied until the circuit is balanced. The circuit is
balanced once the variable impedances cancel out the background noise of the equipment,
leaving only the discharges from the test sample. This is an ideal situation, though the circuit is
much more difficult to set up and was thus not attempted due to constraints imposed by the
equipment available.
20
Figure 5-2: Balanced detection circuit. [4]
This left the option of straight detection. Straight detection detects all noise in the circuit including
the test sample, transformer and connections. This makes it imperative to make all of the
connections firm, and all corners and sharp edges covered by a hollow metal sphere. Any edges
plasticine. The edges and corners are where the electric field is strongest, by covering them the
chances of stray noise in the circuit are reduced and it is much easier to obtain a clean signal.
The negative peak was found by placing a sharp point on the ground near the conductor. This
produced negative corona that appeared on the display showing where the negative peak was.
Once the negative peak was found, it was possible to determine where the discharges were
originating.
21
Figure 5-3: Straight detection circuit. [4]
Figure 5-4: Oscilloscope output with negative peak corona at the top of the ellipse and the
22
Measurements were taken at these voltages as the voltage ascended and then as they
descended. The 13kV reading for the descent was first raised to 14kV before being measured.
The reason for taking measurements as the voltage both increased and decreased was because
partial discharges remain at lower voltages levels than at what they are originally induced at. By
By inputting a step voltage of 100V into the detector, the magnitude of the partial discharge could
be measured by comparing the sizes of their respective lines on the partial discharge detector.
Then by adjusting the attenuation switches on the detector, it was possible to decrease the
magnitude of the step input until it was the same size as that of the discharge. From the value of
the attenuation switches it was then possible to calculate the size of the discharge from the
following formula.
qx = EqCq(1+Cx /Cb)
? qx = EqCq
Cq = 2pF
-dB/20
? qx = 200 x 10 pC
The final test was to measure the Electro Magnetic Field. Again, there were no Australian
Standards to follow for the measurement of the electric field. This test involved placing a Gauss-
Maus in various positions and raising the line to rated voltage. The detector was placed
perpendicular to the way the current was flowing. Though there was no load, this method still
measurement was taken with no voltage on the line and then at rated voltage. The difference
23
between these two readings was the amount of electric field produced by the conductor. This is a
necessary step as there is ambient electric fields from such things as nearby equipment and even
fluorescent lights. These values were then compared with each other and then compared to those
Figure 5-5: The eight positions used for measuring the electric field.
The electric field surrounding the insulator and the conductor were modelled using the computer
program QuickField. This program allows two-dimensional CAD drawings to be imported into it.
Once the drawings have been imported, the dielectric constants of all of the materials can be set.
QuickField then uses Finite Element Method analysis to draw the electric field contours
24
This Finite Element technique places a mesh of many triangles inside all of the areas of the
drawing. As larger triangles do not fit model curves effectively, an approximation by many smaller
triangles is made. The smaller the triangles, the better the approximation of the drawing, but a
much larger problem is the consequence. It is preferable to have good quality triangles, as close
to balanced (equilateral) as possible. A measure of the quality of a triangle is thus the ratio
The version of QuickField used was an evaluation version that was only able to process two
hundred nodes. This meant the mesh incorporated large triangles that reduced accuracy. It
follows that the results are a rough estimate. The difference between a mesh with large spacing
and one with small spacing can be seen below. It is possible to get around this problem in two
ways. One method is to use different mesh grades depending on the curvature of boundary in
that region. This can be done in two ways, either an automatic grading is possible (but not
available), or individual regions can be hand separated and a mesh grade supplied, but this
creates the extra non-trivial task of connecting artificial boundary nodes and may introduce
artificially high curvature. To circumvent the size restrictions, a close-up of the conductor region
was examined. While this provides a local picture, it covers such a small region that it does not
necessarily reflect the global picture. The use of both a wide view and a close view together will
25
Figure 5-6: A Tie Top insulator drawn using a spacing of 1 (left) and 50 (right).
The values for the dielectric constants used in the computer modelling were obtained through the
use of CES Selector V3.1. The value for the AAAC conductor was not available, but as it is a
conductor, it was assumed that the dielectric constant would be extremely high, therefore the
value of 1000 was chosen. All of these values are compared to the value of air. The values for the
It was assumed that the base of the insulators and the pin of the pin post insulators were at 0V,
as the concrete poles the trident structure is used with are fairly conductive. Even if wooden poles
are used, the voltage at the base of the insulator will still be fairly close to earth potential when
26
Table 5-1: Dielectric constants of the materials used in Finite Element analysis.
Porcelain 6 7
Air 1 1
Concrete 8 12
27
Chapter 6
Results
This chapter provides tables and graphs of the data collected during the experiments. The data
presented here for the EMF and the partial discharge is the average for each test, the raw data
can be viewed in Appendix A and Appendix B. Photos of the ERA Discharge Display’s
28
6.1 Leakage Current
At 5.25kV At 6.35kV
At 7.57kV
Stripped 37.5
Stripped 36.25
Stripped 58.2
Bare 57.7
29
Leakage Current vs Voltage
60
55
50
Leakage Current (uA)
45 Clamp Covered
Clamp Stripped
Tie Covered
40 Tie Stripped
Pin Post Covered
Pin Post Stripped
35 Pin Post Bare
30
25
20
5 5.5 6 6.5 7 7.5 8
Voltage (kV)
30
6.2 Electric Field Measurements
31
Side Tie - Covered Conductor Thick
32
Pin Post - Covered Conductor Thick
33
Pin Post - Bare Conductor
1.6
1.4
1.2
1
EMF (mG)
Covered Conductor
0.8
Stripped Conductor
0.6
0.4
0.2
0
0 1 2 3 4 5 6 7 8 9
Position Number
Figure 6-2: Average EMF Measurements for the Clamp Top insulator
34
Average EMF Measurements for the Side Tie Insulator
2.5
1.5
EMF (mG)
Covered Conductor
Stripped Conductor
0.5
0
0 1 2 3 4 5 6 7 8 9
Position Number
Figure 6-3: Average EMF Measurements for the Side Tie Insulator
2.5
1.5
EMF (mG)
Covered Conductor
Stripped Conductor
Bare Conductor
1
0.5
0
0 1 2 3 4 5 6 7 8 9
Position Number
Figure 6-4: Average EMF Measurements for the Pin Post Insulator
35
6.3 Partial Discharge
Ascending Descending
Voltage (kV) Charge (pC) Charge(pC) Charge(pC) Charge (pC) Charge(pC) Charge(pC)
Ascending Descending
Voltage (kV) Charge (pC) Charge(pC) Charge(pC) Charge (pC) Charge(pC) Charge(pC)
Ascending Descending
36
Side Tie - Stripped Conductor
Ascending Descending
Voltage (kV) Charge (pC) Charge(pC) Charge(pC) Charge (pC) Charge(pC) Charge(pC)
Ascending Descending
Voltage (kV) Charge (pC) Charge(pC) Charge(pC) Charge (pC) Charge(pC) Charge(pC)
Ascending Descending
Voltage (kV) Charge (pC) Charge(pC) Charge(pC) Charge (pC) Charge(pC) Charge(pC)
37
Pin Post - Bare Conductor
Ascending
Descending
No Conductor
Bare A
6.35 1
10 6.4
13 7.1
38
Partial Discharge Clamp Top - Covgered Conductor Thick
10
6
Charge (pC)
Ascending
5
Descending
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
12
10
8
Charge (pC)
Ascending
6
Descending
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
39
Partial Discharge Side Tie - Covered Conductor Thick
12
10
8
Charge (pC)
Ascending
6
Descending
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
12
10
8
Charge (pC)
Ascending
6
Descending
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
40
Partial Discharge Pin Post - Covered Condcutor Thick
4
Charge (pC)
Ascending
3
Descending
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
3.5
2.5
Charge (pC)
2
Ascending
Descending
1.5
0.5
0
5 6 7 8 9 10 11
Voltage (kV)
41
Partial Discharge Pin Post - Bare Conductor
3.5
2.5
Charge (pC)
2
Ascending
Descending
1.5
0.5
0
5 6 7 8 9 10 11
Voltage (kV)
42
Partial Discharge Clamp Top Ascending - Covered vs Stripped
12
10
8
Charge (pC)
Covered
6
Stripped
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
12
10
8
Charge (pC)
Covered
6
Stripped
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
43
Partial Discharge Side Tie Ascending - Covered vs Stripped
12
10
8
Charge (pC)
Covered
6
Stripped
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
12
10
8
Charge (pC)
Covered
6
Stripped
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
44
Partial Discharge Pin Post Ascending - Covered vs Stripped
4
Charge (pC)
Covered
3 Stripped
Bare
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
4
Charge (pC)
Covered
3 Stripped
Bare
0
5 6 7 8 9 10 11 12 13 14
Voltage (kV)
45
6.4 QuickField Results
46
Figure 6-19: Clamp Top Insulator – Stripped Conductor
47
Figure 6-20: Tie Top Insulator – Covered Conductor Thick
48
Figure 6-21: Tie Top Insulator – Stripped Conductor
49
Figure 6-22: Pin Post Insulator – Covered Conductor Thick
50
Figure 6-23: Pin Post Insulator – Stripped Conductor
51
Figure 6-24: Pin Post Insulator – Bare Conductor
52
Table 6-4
Stripped 4.592
Stripped 5.440
Stripped 23.075
Bare 21.750
53
Chapter 7
Discussion
All of the insulators displayed a tendency for the leakage current to increase proportionally with
the voltage across the insulator. The Pin Post insulators have a leakage current between 55%
and 65% larger than both the Clamp and tie Top insulators. This is due to the actual physical
design of the different insulators. The Pin Post is much smaller and has only two flanges,
compared to the others four. These flanges increase the distance that the leakage current must
travel and therefore increase the impedance. The Clamp Top and the Tie Top insulators have
almost exactly the same leakage current measurements, usually within 5% of each other. This is
because their flange design is almost identical. The clamp is made of metal and therefore
conducts well, and this is attached to almost exactly where the conductor on the Tie Top is
Stripping the conductor on the Clamp Top insulator at 6.35kV results in a 9.1% increase in the
leakage current, where as stripping the conductor on the Tie Top makes only a 0.1% increase.
Even though it is only a slight increase, over the entire 11kV network these few micro-amps lead
to greater losses, and will therefore be less efficient, providing yet another reason to avoid
Once a partial discharge is initiated, it will remain present as long as its voltage remains above
the extinction level. This explains why larger discharges were observed when the voltage was
54
descending. The original discharges were present as well as those initiated at higher voltage
levels that were yet to be extinguished. If the inception voltage is slightly greater than the rated
voltage of a line and the extinction voltage for the discharges formed is below the rated voltage a
problem may occur. The line may not have partial discharges occurring in it or none of significant
size, but under swell conditions, the inception level may be surpassed, causing discharges to
begin occurring. When the line returns to rated voltage, the discharges will remain and may cause
damage over time if large enough and if they remain for long enough. This can be seen in Figure
6-5 to Figure 6-11, where the descending discharges are generally 25% larger, and in some
Figure 6-12 to Figure 6–17 shows that the stripped conductor always produced larger partial
discharges than the covered conductor. Considering that the purpose of stripping covered
conductor is to reduce the number and size of partial discharges, it would seem that this method
does not work. When the stripping was performed, a few scratches were made by accident on the
aluminium conductor. This could be the source of the discharges, and if so, these would not be of
a detrimental nature to the system as they are not touching the covering and therefore not going
to damage it.
With no conductor connected, partial discharges were detected. This was due to background
discharges in the equipment. This can be removed with the use of a balanced detection circuit,
though due to the lack of equipment this was not an option. The levels of discharge detected at
10kV were approximately 6.4pC. AS mentioned earlier, AS 3599.2 recommends that the
maximum allowable discharge at this voltage level is 5pC. As the background noise is already
above this level it is impossible to verify the exact magnitude of the discharges in a conductor if
they are below this background level. If the magnitude is significantly larger than this 6.4pC level,
it is safe to say that these larger discharges are entirely within the conductor. This is due to the
unlikelihood of an already existing discharge in the test equipment being superimposed by a new
55
At 10kV, the entirety of the Clamp Top and Tie Top configurations have discharges in the range
of 5pC to 7pC. This in the same range as the background noise, and due to the inherent
difficulties in measuring the discharge, it seems that the line is not contributing to the size of the
discharges. The only way to validate this assumption is to use balanced detection, which was not
The discharges also had to be measured at 13kV. This test was much more conclusive, as none
of the samples displayed discharges greater than 11.25pC and the maximum allowable discharge
was 50pC. Therefore all of the configurations passed the Australian Standard level.
The Pin Post insulators provided some difficulties when being tested for partial discharges. The
readings were fairly clear at 6.35kV, but as the voltage approached 10kV, a number of sizeable
coronal discharges began occurring regularly. These discharges made it particularly difficult to
make accurate measurements and once the voltage was increased to 13kV it was impossible to
obtain accurate readings due to the amount of corona, except for the covered conductor which
During the testing, a spark gap was placed across two of the terminals on the discharge detector.
This was to reduce the chance of damage to the device as the flashover voltage of 75V
corresponded to the maximum allowable current in the detector. On a few occasions this spark
gap flashed over and produced a great deal of electrical noise. This noise generally occurred at
around 12kV on the secondary side. This flashover voltage was not always the same as when
the partial discharge measurements were taken, no flashover was heard or seen, even up to
14kV. It is possible that the spark gap was on the verge of flashing over and was producing these
great amounts of noise. The spark gap could be removed, though this would leave the detector
56
7.3 QuickField Simulations
A CAD drawing of the object being modelled is also provided, as some of the electric field
diagrams do not show the insulator clearly. Equipotential lines are also shown on the electric field
diagrams. From these lines, it is possible to see that the electric field is the derivative with respect
to distance of voltage. Where the voltage lines are more spread out, the electric field is smaller.
When there is a conductor, like the pin attachment in the pin post insulator configuration, there is
negligible voltage drop across its length; therefore there is no electric field within it.
These diagrams only show a simplified two-dimensional view, but they do provide suitable
information on the magnitudes of the electric fields surrounding the insulator and conductor
connections. Table 6.4 represents the maximum field strength found in each of the electric field
diagrams. The maximum values were generally found near the insulator and conductor
connections. It can be seen that stripping the covering from the conductor does reduce the
magnitude of the electric field in the surrounding area. The reason for the comparatively high
values for the pin post insulator are due to the close proximity of the pin to the live conductor.
There is a potential difference of 6.35kV across the thin top of the insulator. A large potential drop
across a small distance requires a steep voltage gradient, and due to the magnitude of the
electric field is the derivative of the voltage with respect to distance, a large electric field is also
present. The reverse logic can be used to understand why there is very small electric field near
the conductor but away from the insulator. The outside of the conductor is floating and not being
forced to earth potential, therefore the electric field is not required to be as large as the change in
The insulating covering of the conductor has an extremely low dielectric constant of 2.3. This
makes it a very strong insulator and therefore it can withstand a large voltage across a small
distance without flashing over. When two materials of the same thickness with different dielectric
constants are in series with each other with a voltage across them, most of the voltage will be
57
across the material with the lower dielectric value. This causes an extremely high electric field to
be present in the low dielectric material. By removing the insulating covering, the conductor is
now directly connected to the insulator. The porcelain insulators have a dielectric constant of
approximately six, and this allows the voltage to gradually decrease across it and thus reduces
58
7.4 Electric Field
The Pin Post insulator had slightly higher electric field values than the other insulators. This is
due to the large pin inserted in the base of the insulator and matches exactly with the results
Figure 6-2 to Figure6-4 show that stripping the polyethylene covering near where the insulator
attaches almost always reduces the surrounding electric field. This is as expected as the voltage
now is directly applied to the insulator. This insulator has a slightly larger dielectric constant and
is a larger size than the polyethylene covering and therefore the voltage will decrease at a slower
rate, resulting in a smaller electric field. Again, this is the same result as attained through
QuickField.
59
Chapter 8
Conclusions
Stripping the polyethylene covering from near the insulator attachment resulted in a few changes
to the system characteristics. The leakage current increased as the conductor was now in direct
contact with the insulator. The surrounding electric field decreased due to the removal of a low
dielectric material that now allows the voltage to decrease at a reduced rate and thus a reduced
electric field. Finally the amount and size of the partial discharges increases compared to covered
conductor. The theory behind stripping the covering from the conductor is to reduce the leakage
current and the size of the partial discharges by reducing the electric field. In almost all cases the
electric field is successfully reduced, but the leakage current and the occurrence of partial
discharges both increase. Performing the stripping of the conductor is a difficult job requiring
special tools and training. When the conductor is stripped, it also increases the systems
susceptibility to wildlife and vegetation outages. This appears to be the wrong method of
protecting the conductor from damage, as there are no benefits to this practice.
In respect to the insulators, it is difficult to determine a clear leader. The Tie Top and the Clamp
Top are very similar in all respects, while the Pin Post has a higher leakage current but it is
impossible to deduce how they are affected by partial discharges, due to the large background
noise.
This background noise was also a problem for the partial discharge tests undertaken at 10kV, as
the background noise was larger than the acceptable value from the relevant Australian
Standard. The spark gap on the discharge detector is believed to be the cause of the noise, as all
other edges and points were removed from the circuit or covered by metal spheres. This noise
makes it difficult to arrive at a definitive conclusion as to whether the standards are met at 10kV.
60
It seems that stripping the conductor only produces problems without any benefits. The leakage
current for the stripped conductor is greater than that of the covered conductor, as is the amount
of partial discharges. The theory behind stripping the conductor is to reduce the amount and size
of the partial discharges, but this is not the case in practice. The actual task of stripping the
conductor is also a difficult task requiring special tools and training. This would require increased
expenditure for any company wishing to implement the practice of stripping their covered
conductors. Stripping the conductors also increases the likelihood of vegetation and wildlife
reduced outages, decreasing the reliability of the supply to the customers. This thesis shows that
the practice of stripping covered conductors is not advised as it is not only harmful to the
61
Future Plan
This thesis provides a sturdy platform for the study of insulators and their interactions with
covered and stripped conductor. This topic could be further investigated through the analysis of a
wider variety of insulators and conductors. These could then be analysed more precisely with the
implementation of a balanced partial discharge detector circuit. This would allow the background
noise to be removed from the signal and allow the exact location of the source of the discharges
This would enable the field to be modelled more accurately and the field could be viewed from
various angles.
62
References
[1] B. Hart, “HV Overhead Line – The Scandinavian Experience,” Power Engineering Journal, Vol.
Insulator/Tie Configurations,” IEEE Transactions on Power Delivery, Vol. PWRD-1, No. 4, Oct.
[3] D.A. Swift, “Electrical Puncture of the Insulating Sheath of Covered Overhead Power-Line
[4] F.H. Kreuger, Partial Discharge Detection in High-Voltage Equipment, Temple Press, England,
1989.
[5] J. Roughan and G. Dowling, Development of Fittings for Covered Conductors (CCT), Systems
[6] K.A. Gosden, Covered Conductor (CCT) Implementation, ELECTRO Technical Consultants,
1998.
[7] Electricity Trust of South Australia, Overhead Insulated Systems in south Australia, Electricity
[8] ELECTRO Technical Consultants, Insulated HV Overhead Mains – Technical and Economic
63
[9] Australian Standard, AS/NZS3599.2 Electric cables – Aerial bundled – Polymeric insulated –
64
Bibliography
S. Davis, The Lightning Performance of the Overhead 11kV Energex Trident Structures, thesis,
S.R. Krishnamurthy and P. Selvan, “Use of AAAC in a Distribution Network – A Strategy for
Energy and Cost Reduction,” Power Engineering Journal, Vol. 9, No. 3, Jun. 1995, pp 133-136.
E. Kuffel and W.S. Zaengl, High Voltage Engineering: Fundamentals, Pergamon Press, Australia,
1984.
J.W. McAuliffe, Hendrix Aerial Spacer Cable System an Option for System Reliability
65
Appendix A – EMF Results
CCT A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.5 0.9 0.4
1.39 0.4 0.9 0.5
1st Left 1.05 0.45 1 0.55
1.39 0.35 1.15 0.8
2nd Left 1.05 0.5 0.8 0.3
1.39 0.3 0.95 0.65
High Post 2.15 0.55 1.5 0.95
Low Post 1.93 0.6 1.9 1.3
CCT B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 1.1 0.7
1.39 0.3 1.2 0.9
1st Left 1.05 0.4 0.8 0.4
1.39 0.3 1.1 0.8
2nd Left 1.05 0.4 0.8 0.4
1.39 0.3 1.05 0.75
High Post 2.15 0.6 1.75 1.15
Low Post 1.93 0.6 1.6 1
CCT C
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.9 0.5
1.39 0.4 1.1 0.7
1st Left 1.05 0.4 0.8 0.4
1.39 0.4 1 0.6
2nd Left 1.05 0.4 0.7 0.3
1.39 0.3 1 0.7
High Post 2.15 0.6 1.9 1.3
Low Post 1.93 0.5 2.3 1.8
CCT D
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 0.8 0.35
1.39 0.3 1.4 1.1
1st Left 1.05 0.4 0.85 0.45
1.39 0.4 1 0.6
2nd Left 1.05 0.5 0.7 0.2
1.39 0.35 1.1 0.75
High Post 2.15 0.5 2.1 1.6
Low Post 1.93 0.5 1.8 1.3
66
Side Tie - Covered Conductor Thick
CCT A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 0.9 0.45
1.39 0.4 1 0.6
1st Left 1.05 0.45 0.85 0.4
1.39 0.35 0.9 0.55
2nd Left 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
High Post 2.15 0.6 2.55 1.95
Low Post 1.93 0.5 2.3 1.8
CCT B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.7 0.3
1.39 0.3 1 0.7
1st Left 1.05 0.4 0.8 0.4
1.39 0.3 1 0.7
2nd Left 1.05 0.4 0.8 0.4
1.39 0.3 1.1 0.8
High Post 2.15 0.6 3.05 2.45
Low Post 1.93 0.6 2.2 1.6
CCT C
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.8 0.4
1.39 0.35 1.2 0.85
1st Left 1.05 0.4 0.7 0.3
1.39 0.35 1 0.65
2nd Left 1.05 0.4 0.65 0.25
1.39 0.3 0.7 0.4
High Post 2.15 0.6 2.6 2
Low Post 1.93 0.6 1.9 1.3
CCT D
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.7 0.3
1.39 0.3 1.1 0.8
1st Left 1.05 0.4 0.7 0.3
1.39 0.3 1 0.7
2nd Left 1.05 0.4 0.5 0.1
1.39 0.3 0.9 0.6
High Post 2.15 0.65 2.35 1.7
Low Post 1.93 0.7 1.8 1.1
67
Pin Post - Bare Conductors
Bare A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.5 0.8 0.3
1.39 0.3 1.1 0.8
1st Left 1.05 0.4 0.8 0.4
1.39 0.3 0.9 0.6
2nd Left 1.05 0.4 0.7 0.3
1.39 0.35 0.8 0.45
High Post 2.15 0.55 3.15 2.6
Low Post 1.93 0.5 2.95 2.45
Bare B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 0.9 0.45
1.39 0.35 1.2 0.85
1st Left 1.05 0.45 0.8 0.35
1.39 0.4 1.1 0.7
2nd Left 1.05 0.5 0.7 0.2
1.39 0.4 0.8 0.4
High Post 2.15 0.6 2.55 1.95
Low Post 1.93 0.55 3.2 2.65
Bare C
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.5 1 0.5
1.39 0.4 1.1 0.7
1st Left 1.05 0.5 0.9 0.4
1.39 0.4 1.3 0.9
2nd Left 1.05 0.5 0.9 0.4
1.39 0.4 1.3 0.9
High Post 2.15 0.65 2.55 1.9
Low Post 1.93 0.55 2.95 2.4
Bare D
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 1.2 0.75
1.39 0.35 1.4 1.05
1st Left 1.05 0.45 0.8 0.35
1.39 0.35 1.15 0.8
2nd Left 1.05 0.5 0.7 0.2
1.39 0.35 0.7 0.35
High Post 2.15 0.6 2.25 1.65
Low Post 1.93 0.55 2.05 1.5
68
Pin Post - Covered Conductors
CCT A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 1.05 0.6
1.39 0.3 1.3 1
1st Left 1.05 0.45 0.8 0.35
1.39 0.3 0.85 0.55
2nd Left 1.05 0.4 0.6 0.2
1.39 0.3 0.8 0.5
High Post 2.15 0.6 1.85 1.25
Low Post 1.93 0.6 1.85 1.25
CCT B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.45 1.1 0.65
1.39 0.3 1.15 0.85
1st Left 1.05 0.45 0.9 0.45
1.39 0.35 1.1 0.75
2nd Left 1.05 0.4 0.8 0.4
1.39 0.3 0.7 0.4
High Post 2.15 0.6 2.15 1.55
Low Post 1.93 0.7 1.85 1.15
CCT C
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 1 0.6
1.39 0.3 1.15 0.85
1st Left 1.05 0.4 0.8 0.4
1.39 0.3 0.9 0.6
2nd Left 1.05 0.4 0.75 0.35
1.39 0.3 1.15 0.85
High Post 2.15 0.6 3.05 2.45
Low Post 1.93 0.55 2.85 2.3
CCT D
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 1.1 0.7
1.39 0.3 1.5 1.2
1st Left 1.05 0.4 0.9 0.5
1.39 0.3 1.2 0.9
2nd Left 1.05 0.4 0.75 0.35
1.39 0.3 1.1 0.8
High Post 2.15 0.7 3.3 2.6
Low Post 1.93 0.6 3.4 2.8
69
Pin Post - Stripped Conductor
Stripped A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.9 0.5
1.39 0.35 1.05 0.7
1st Left 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
2nd Left 1.05 0.4 0.7 0.3
1.39 0.3 1 0.7
High Post 2.15 0.7 2.6 1.9
Low Post 1.93 0.6 2 1.4
Stripped B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
1st Left 1.05 0.4 0.8 0.4
1.39 0.3 0.8 0.5
2nd Left 1.05 0.4 0.7 0.3
1.39 0.3 0.8 0.5
High Post 2.15 0.6 2.25 1.65
Low Post 1.93 0.6 2.4 1.8
Stripped A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
1st Left 1.05 0.4 0.85 0.45
1.39 0.3 1.1 0.8
2nd Left 1.05 0.4 0.65 0.25
1.39 0.3 0.8 0.5
High Post 2.15 0.5 1.55 1.05
Low Post 1.93 0.5 1.75 1.25
Stripped B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.5 0.8 0.3
1.39 0.35 1.1 0.75
1st Left 1.05 0.5 0.7 0.2
1.39 0.4 1.1 0.7
2nd Left 1.05 0.5 0.65 0.15
1.39 0.4 0.95 0.55
High Post 2.15 0.65 1.45 0.8
Low Post 1.93 0.65 1.85 1.2
70
Side Tie - Stripped Conductor
Stripped A
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
1st Left 1.05 0.4 0.7 0.3
1.39 0.3 1 0.7
2nd Left 1.05 0.45 0.6 0.15
1.39 0.3 1.1 0.8
High Post 2.15 0.65 2.2 1.55
Low Post 1.93 0.6 1.5 0.9
Stripped B
Position Height (m) Ambient Gauss (mG) Total Gauss (mG) Cable Contribution (mG)
Centre 1.05 0.4 0.8 0.4
1.39 0.3 0.95 0.65
1st Left 1.05 0.4 0.7 0.3
1.39 0.3 0.9 0.6
2nd Left 1.05 0.5 0.65 0.15
1.39 0.3 0.9 0.6
High Post 2.15 0.65 1.55 0.9
Low Post 1.93 0.65 1.25 0.6
71
Appendix B – Partial Discharge Results
CCT C
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 47 0.89 41 to 37.9 14 to 13 26 10.02
29.0 10 34 3.99 to 29.0 to 10 31 5.64
37.9 13 27 8.93 to 18.5 to 6.35 45 1.12
CCT D
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 45 1.12 41 to 37.9 14 to 13 27 8.93
29.0 10 31 5.64 to 29.0 to 10 30 6.32
37.9 13 27 8.93 to 18.5 to 6.35 43 1.42
CCT C
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 37 2.83 41 to 37.9 14 to 13 26 10.02
29.0 10 30 6.32 to 29.0 to 10 30 6.32
37.9 13 27 8.93 to 18.5 to 6.35 36 3.17
CCT D
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 45 1.12 41 to 37.9 14 to 13 26 10.02
29.0 10 30 6.32 to 29.0 to 10 29 7.10
37.9 13 27 8.93 to 18.5 to 6.35 40 2.00
Stripped A
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 39 2.24 41 to 37.9 14 to 13 26 10.02
29.0 10 32 5.02 to 29.0 to 10 30 6.32
37.9 13 26 10.02 to 18.5 to 6.35 37 2.83
72
Stripped B
Ascending Descending
Volta ge (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 39 2.24 41 to 37.9 14 to 13 25 11.25
29.0 10 31 5.64 to 29.0 to 10 29 7.10
37.9 13 25 11.25 to 18.5 to 6.35 36 3.17
Stripped A
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 39 2.24 41 to 37.9 14 to 13 26 10.02
29.0 10 29 7.10 to 29.0 to 10 28 7.96
37.9 13 26 10.02 to 18.5 to 6.35 36 3.17
Stripped B
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 39 2.24 41 to 37.9 14 to 13 27 8.93
29.0 10 30 6.32 to 29.0 to 10 29 7.10
37.9 13 25 11.25 to 18.5 to 6.35 37 2.83
CCT C
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 49 0.71 41 to 37.9 14 to 13 30 6.32
29.0 10 38 2.52 to 29.0 to 10 38 2.52
37.9 13 30 6.32 to 18.5 to 6.35 47 0.89
CCT D
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 50 0.63 41 to 37.9 14 to 13 32 5.02
29.0 10 40 2.00 to 29.0 to 10 40 2.00
37.9 13 34 3.99 to 18.5 to 6.35 49 0.71
73
Pin Post - Stripped Conductor
Stripped A
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 47 0.89 41 to 37.9 14 to 13 n/a
29.0 10 37 2.83 to 29.0 to 10 35 3.56
37.9 13 n/a to 18.5 to 6.35 46 1.00
Stripped B
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 47 0.89 41 to 37.9 14 to 13 n/a
29.0 10 39 2.24 to 29.0 to 10 37 2.83
37.9 13 n/a to 18.5 to 6.35 45 1.12
Bare A
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 49 0.71 41 to 37.9 14 to 13 n/a
29.0 10 39 2.24 to 29.0 to 10 35 3.56
37.9 13 n/a to 18.5 to 6.35 47 0.89
Bare B
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 47 0.89 41 to 37.9 14 to 13 n/a 200.00
29.0 10 35 3.56 to 29.0 to 10 35 3.56
37.9 13 n/a 200.00 to 18.5 to 6.35 45 1.12
Bare C
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 49 0.71 41 to 37.9 14 to 13 n/a
29.0 10 39 2.24 to 29.0 to 10 37 2.83
37.9 13 n/a to 18.5 to 6.35 46 1.00
Bare D
Ascending Descending
Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC) Voltage (V) Voltage (kV) Attenuation (dB) Charge (pC)
18.5 6.35 49 0.71 41 to 37.9 14 to 13 n/a
29.0 10 40 2.00 to 29.0 to 10 38 2.52
37.9 13 n/a to 18.5 to 6.35 47 0.89
74
Appendix C – Partial Discharge Oscilloscope Photographs
75
Photograph of partial discharges with ellipse open of Covered Conductor Thick on a Tie
Photograph of partial discharges with ellipse open of Stripped Conductor on a Tie Top
insulator at 13kV.
76
Photograph of partial discharges with ellipse open of Covered Conductor Thick on a
Photograph of partial discharges with ellipse open of Stripped Conductor on a Clamp Top
insulator at 13kV.
77
Photograph of partial discharges with ellipse open of Covered Conductor Thick on a Pin
Photograph of partial discharges with ellipse open of Stripped Conductor on a Pin Post
insulator at 13kV. Note that it is impossible to see the 100V input marker and hence
78
Photograph of partial discharges with ellipse open of Bare Conductor on a Pin Post
insulator at 13kV. Note that it is impossible to see the 100V input marker and hence
79